Background
Arterials carry heavy commuter traffic; independent signals at adjacent intersections often cause a "red-wave". Green-wave coordination lets platoons pass continuously at the design speed, yet demand fluctuation and turning flows erode the benefit.
Objective
Design bi-directional green-wave timing under given geometry and demand to meaningfully reduce peak-hour delay and stops.
Method
Built a 6-intersection arterial in SUMO; used Webster method for cycle & split, then a MAXBAND-style graphical method for offset; TraCI collected queue & speed in real time to iteratively tune bandwidth.
Results & Analysis
24h 流量曲线
速度分布
密度热力图
Interactive Demo
Reflection
This project taught me that green-wave is a dynamic system rather than a static timing table—turning flows and pedestrian signals significantly perturb the bandwidth. Next step: online offset tuning via reinforcement learning.